Saudi Insurance Authority and HUMAIN sign AI cooperation agreement

The Insurance Authority signed a cooperation agreement with HUMAIN on the sidelines of LEAP 2026 in Riyadh, opening a formal channel to explore how the Public Investment Fund's artificial intelligence company might support Saudi Arabia's insurance sector. The agreement is explicitly exploratory. It covers three things: examining HUMAIN's AI compute, products, services and wider technology ecosystem for potential use inside insurance regulation and the industry it oversees; establishing a route for secure access to insurance-sector data and to HUMAIN's AI products and services in support of the Authority's strategic priorities; and identifying priority use cases where AI might be applied across the sector. None of that is a deployment, a licensing arrangement or a data-sharing agreement in the operational sense — it is a mandate to go and look. That distinction matters more in insurance than in most sectors regulators touch. An insurance regulator's core job is protecting policyholders and keeping insurers solvent, which means its data is unusually sensitive: claims histories, underwriting models, reserve calculations, and the loss data that determines whether a company can pay out when it needs to. Any arrangement that touches that data has to satisfy far stricter scrutiny than a typical government IT modernisation project, which is presumably why the agreement's language is confined to "exploring" and "enabling secure access" rather than committing to a specific system or timeline. HUMAIN's pitch to a regulator like the Insurance Authority is the same one it has been making to companies across the Saudi public and private sector all week: compute, models and a technology ecosystem that HUMAIN wants domestic institutions to build on rather than import piecemeal from abroad. SDAIA and HUMAIN opened an AI procurement framework spanning eleven government ministries earlier this year , and the Insurance Authority's agreement reads as an extension of that same logic into a regulator with its own supervisory mandate rather than a service ministry. Insurance is a plausible sector for AI to matter quickly. Underwriting, fraud detection and claims processing are all pattern-recognition problems sitting on large historical datasets, and insurers globally have been among the more aggressive adopters of machine learning for exactly that reason. A regulator that understands what its supervised entities are doing with AI — and has the technical capacity to evaluate model risk rather than take a vendor's word for it — is in a stronger position than one that is purely reactive. What the agreement does not specify is any of the operational detail that would make it possible to judge scale or timeline: no named use case has been selected, no data-sharing protocol has been published, and no date has been set for when exploration might turn into an actual system. HUMAIN has signed a wide range of exploratory and cooperation agreements across LEAP 2026, from a data platform partnership with MinIO to its investment in enterprise AI firm MOZN , and this one sits at the more preliminary end of that spectrum. The test is whether the Insurance Authority publishes a specific use case with an actual timeline, rather than a renewed cooperation agreement a year from now. TechScoop is reporting from LEAP 2026 in Riyadh.